General configuration management of large systems typically involves comparing objects, such as host computer systems or virtual machines, against a reference object. Objects that are originally configured to be similar to the reference object may change over time, resulting in objects that differ from the reference object. Objects under configuration management that differ from the reference object are identified and corrected, bringing such objects back into similarity with the reference object. Similarly, objects subject to a regulatory compliance policy may be out of compliance with that policy, exposing the entity responsible for the objects to potentially large fines or other liability. Objects subject to such regulations that differ from the reference policy are identified and corrected, bringing objects into compliance with the regulatory policy.
One approach to identifying objects that differ from the reference object is to tabulate the properties of each object against the properties of the reference object. A configuration manager visually scans the tabulated properties, looking for objects that differ from the reference object. One drawback with this approach is that the size of the tabulated properties becomes quite large when the objects to compare with the reference object number in the hundreds or the thousands. Visual comparison of the objects in such large systems is time consuming and prone to human error. Consequently, objects that are out of conformity or compliance may not be identified, resulting in reduced system performance or increased liability.